A new structural optimization method, which is based on subset simulation optimization (SSO) and the Lagrangian multiplier method, is proposed to solve structural optimization problems under constraints in this paper. The Lagrangian multiplier method is used to handle multiple constraints. Then, SSO is used to solve the transformed unconstrained problem and search for the global optimum. In order to guarantee the robustness of convergence, the updating method of penalty factor is modified for this purpose. The accuracy, robustness and efficiency of the proposed method are demonstrated by two examples. The optimization results of proposed method are compared with those obtained by other optimization methods available in the literature. It indicates that the proposed method has excellent performance on searching for the global optimum.